Benefits of Redundancy & Cold Spares

Importance of Equipment Redundancy

Downtime is the true enemy of every network. Downtime is frustrating, destructive, and worst of all, costly. Businesses can rapidly lose money with even small amounts of network downtime, so everyone wants ways to avoid it.

There really is one answer: redundancy.

What Is Redundancy?

At a technical level, redundancy is the repetition of something, but when it comes to logistics — and more importantly network design — redundancy is a design practice. A redundant system can never break from a single point of failure.

So, if a single networking device goes down, the network as a whole will still function.

It’s a complicated aspect of network design, but it’s essential for avoiding devastating downtime.

When it comes to networking redundancy, there are two primary concerns: connectivity and power. Connective redundancy ensures that communication can succeed even when a networking node goes down. Power redundancy takes the same look at power supplies for the network, so unless an entire region loses power, a redundant system won’t fail if, say, a single breaker is flipped.

What Are Cold Spares?

In pursuit of redundancy, it’s essential to have extra parts and equipment. The extras allow you to fix problems as they arise, and that brings up the point of cold spares.

Cold spares are spare parts or devices that are not currently deployed in the system. As an example, you might have an extra switch that is sitting on a shelf, completely unused. If a switch in the network goes down, you can grab a cold spare and install it.

Since the spare was already on hand, downtime is minimized and problems can be resolved quickly.

How Many Cold Spares Do You Need?

Of course, cold spares cost money, and you always have to think about how to be cost-effective with your network designs. Redundancy is valuable, but it also makes the network more expensive. So, how many cold spares do you need?

That ultimately depends on how you design the network. Ideally, you will have a cold spare available for every critical component of the network, but a simple design choice can make this easier.

If you standardize your equipment as much as possible, a smaller number of cold spares can serve as redundant spares for a large portion of the whole. As an example, if you have 10 different nodes that all use the same model of networking switch, then a single cold spare can serve as backup for all 10.

This is based on the assumption that not all of the nodes will fail simultaneously, but you get the idea.

What About Hot Spares?

While we’re on the topic of cold spares, it also makes sense to discuss warm and hot spares.

If a cold spare is just sitting around on a shelf, then a warm spare is pretty similar. Warm spares are still not already deployed on the network, but they are preconfigured. That way, when they are needed, they can be deployed more quickly.

Naturally, warm spares come with labor costs, since you’re configuring them before they are used. Additionally, warm spares are often less universal. A cold spare might be able to back up a wider range of equipment than a preconfigured warm spare.

Hot spares are very different. These devices are already installed in the network. If a device goes down, the system automatically switches to the hot redundancy.

This is common in networking switch stacks. If the primary controller of the stack goes down, a backup is already programmed and attached and ready to take over the management of the stack. You can fix the primary controller, but until you do, the network continues to function without any downtime at all.

Additional Learning Center Resources